The broad goal of our research program is to understand the biological basis of language. In this Project, Origins of Communication Disorders in Williams Syndrome Infants, we address this important issue from a new perspective. We propose a series of studies of the origins of language and its relation to cognition and affect in children with a rare metabolic disorder, Williams Syndrome (WMS). Our pilot studies of WMS adolescents point to a unique behavioral profile in which there is a striking fractionation of higher cortical functions: linguistic abilities appear to be selectively preserved in the face of severe general cognitive deficits. Our preliminary studies of WMS infants indicate extreme retardation in all developmental milestones, including language. Almost nothing is known about the initial capacity of WMS; thus we propose to examine the origins of language and cognition of WMS and matched Down Syndrome (DNS) infants between 1 and 6 years of age, in order to test alternative theories of language, cognition and brain organization. The specific hypotheses to be tested fall under the following categories: (1) The Decoupling of Language from Cognition in the Development of WMS Infants. We examine cognitive and communicative prerequisites to language, to assess the hypothesis that particular linguistic abilities are dependent on the development of specific cognitive functions, or alternatively, that the two domains are dissociable. (2) Differential Impairment of Components of Language. Syntactic functions in older WMS are remarkably spared, but semantic organization appears to be deviant; WMS show a proclivity for atypical words. We examine the basis for this dissociation in the origins of language in WMS, in lexical development and the emergence of grammar. (3) Dissociations within Domains of Cognition. Even within the domain of spatial cognition, there are clusters of sparing and impairment which appear to be specific to WMS: markedly impaired visuospatial skills but a remarkable capacity for facial recognition. Moreover, WMS spatial deficit reveals selective attention to details of a configuration at the expensive of a whole, whereas DNS results in the opposite profile. We will investigate the basis for this dissociation, and explore the possibility that the unusual profile observed in visual-spatial cognition is related to the relative sparing of language in older WMS. (4) Neural Substrate for Language in WMS. These studies will be carried out in conjunction with studies of the neural correlates of language (Project 5), to address specific questions about the neural substrate for this unusual profile in language and non-linguistic cognition. Because WMS presents a rare dissociation of language from other cognitive capacities, it provides an unusual opportunity to explore one of the central issues of developmental cognitive neuroscience, i.e. the relative autonomy of language and/or the dependence of language on other mental/neural systems. These studies will also contribute to our understanding of brain organization for language.